跟踪从月份开始日期起的周间隔

Question

我试图找出从 2021 年 1 月 1 日开始，每 1 周有多少跑步者报名参加。我的查询没有问题，但我无法使用日期达到 7 天数字标记的 mod (%) 来完成我的查询。

---My Schema
CREATE TABLE runners (
  "runner_id" INTEGER,
  "registration_date" DATE
);
INSERT INTO runners
  ("runner_id", "registration_date")
VALUES
  (1, '2021-01-01'),
  (2, '2021-01-03'),
  (3, '2021-01-08'),
  (4, '2021-01-15');


runner_id   registration_date
1   2021-01-01
2   2021-01-03
3   2021-01-08
4   2021-01-15

所需的 O/P：

start_of_week   signups
2021-01-01    2
2021-01-08    1
2021-01-15    1

---My Query

WITH data1 AS (
  SELECT
    runner_id,
    registration_date,
    REGISTRATION_DATE - (DATEDIFF(DAY, REGISTRATION_DATE, '2021-01-01' ) % 7)  AS start_of_week
  FROM runners
)
select start_of_week, count(runner_id) as signup from data1 group by start_of_week;

错误

操作数类型冲突：日期与 int 不兼容

我也尝试将日期格式切换为带有 7 的 mod，但没有奏效。

 registration_date - ((registration_date - '2021-01-01') % 7)

错误

数据类型 date 和 varchar 在减法运算符中不兼容。

Answer 1

根据 cmets，试试这个：

 WITH data1 AS (
  SELECT
    runner_id,
    registration_date,
    (datediff(DAY,  '2021-01-01',REGISTRATION_DATE ) / 7) + 1 AS start_of_week
  FROM runners
)
select
  dateadd(week, start_of_week - 1, '2021-01-01'),
  count(runner_id) as signup
from data1
group by start_of_week;

dbfiddle

https://dbfiddle.uk/?rdbms=sqlserver_2019&fiddle=91a00cf6223f3c1367b6e8f014dccd6f

如果您想摆脱时间，也可以将其转换为日期

select
  cast(dateadd(week, start_of_week - 1, '2021-01-01') as date),
  count(runner_id) as signup
from data1
group by start_of_week;